Effect of long-acting injectable antipsychotics on emergency department visits and hospital admissions in people with bipolar disorder: A retrospective mirror-image analysis from the Northern Milan Area Cohort (NOMIAC) study

Background: Mood recurrences in bipolar disorder (BD) are often associated with poor treatment adherence. Despite long-acting injectable antipsychotics (LAIs) may favor treatment compliance, their use in BD is still poorly explored. Methods: This mirror-image study investigated the effect of LAIs initiation on the number of emergency department (ED) visits and days of hospitalization, among individuals with BD from the mental health services of a large area of the Metropolitan City of Milan. The mirror periods were 365 days either side of the LAI initiation. Individual medical records were retrospectively reviewed. Results: Sixty-eight individuals with BD initiating a LAI over the index period were included. We estimated that LAI initiation overall reduced both ED visits (p = 0.002) and days of hospitalization (p < 0.001). This remained true only for those participants who i) continued LAI for the entire 12-month period of observation and ii) were treated with a second-generation antipsychotic LAI. In addition, LAI initiation reduced number of hospitalization days during hypo/manic (p = 0.013), but not depressive (p = 0.641) episodes, as well as compulsory admission days (p = 0.002). Limitations: Due to the retrospective design, we could not collect systematic information on symptom severity and reasons of LAI discontinuation. Moreover, the limited sample size did not allow us to estimate effectiveness of single LAI agents. Conclusions: Our study provides additional insight on the effectiveness of LAIs in BD, supporting their clinical utility for pragmatic outcomes such as ED visits and hospitalizations. Further longitudinal research is needed to clarify the real-world effectiveness of LAIs for BD clinical management.